Weekly digest: what’s happening in open science?

Steph Macdonald

Featuring highlights from the European Meeting of ISMPP, discrepancies in disclosure reporting and the time implications of introducing a data availability statement.

Open access highlights from ISMPP via ISMPP

Following the announcement of Ipsen’s mandatory open access policy on 22 January 2019, will other industry contenders soon follow suit? This year’s European Meeting of the International Society for Medical Publications Professionals (ISMPP), which took place in London last week, shed light on the future of open access in industry-funded research. After transitioning to mandatory open access last year, representatives from Shire (now part of Takeda) announced that, between January and December 2018, 91% of Shire research papers were published open access. Over one-third of these were published under a CC BY licence, making the research articles free to read, share and use. Further support for open access publishing came from a joint poster presentation by Oxford PharmaGenesis and Shire. This focused on the finding that, although results from clinical trials are published in both open access and subscription journals, the number of citations and social media shares were higher for papers published open access.

Plan S was also a prominent talking point, with David Sweeney from Research England delivering guidance on the implementation of the plan. In addition to talking through the details of the plan, David offered reassurance about the hotly debated issues emerging from the presence of article processing charges (APCs), with the promise that cOAlition S will be commissioning a study to ensure that APCs are fair and reflective of the impact of each journal. This presentation was followed by a talk from Claire Moulton from the Company of Biologists, which highlighted some of the concerns shared by many non-profit community journals about Plan S. These included the tight timelines and the anti-hybrid views of the plan, as well as the lack of transparent reporting of APCs by journals.

Finally, Andy Powrie–Smith, Executive Communications Director of the European Federation of Pharmaceutical Industries and Associations, delivered an interesting presentation on how open access could help address ‘fake news’ and the difficulty of balancing complex data with attention-grabbing headlines.

Financial disclosures in medical publishing via The Publication Plan

To ensure full transparency, publishers require authors to report disclosures, including any financial relationships between healthcare companies and authors. Articles published in The New York Times and ProPublica last December highlighted that some leading medical figures are not disclosing this information, despite receiving large payouts from drug companies for various activities. A closer examination as to why these are not being disclosed by all researchers showed that the journals themselves could be to blame. Guidelines between journals are inconsistent and confusing, and lack clarity on what constitutes a financial relationship and how much detail authors must report. Furthermore, journals do not perform adequate checks on author claims, despite some information being relatively easy to find through online databases. In a bid to improve disclosure reporting, platforms such as Convey and standardized disclosure forms have been created. However, ultimately, journals should play a stronger role in guiding authors, through clear advice on what relationships should be disclosed.

Data sharing via The Scholarly Kitchen

Data availability statements (DASs) provide information about where data supporting publication results are reported. Publishers, institutions and funding agencies are increasingly using mandatory DASs to ensure that data is discoverable and that they can be shared and reused, thus aiding scientific advancement and global academic collaboration. However, there are concerns regarding the potential cost implications arising from the introduction of a mandatory DAS, because of the additional time required by editors and production staff to ensure accuracy. To investigate this, a study involving the analysis of 557 manuscript submissions was conducted across participating Nature Publishing Group journals. Its results showed that the inclusion of a mandatory DAS in all accepted articles added approximately 15–20 minutes to editorial and production time per paper. Standardized data policies were introduced two years ago by Springer Nature, and, since then, Elsevier, Wiley, Taylor & Francis, Hindawi and The BMJ have adopted similar practices, indicating that the importance of data sharing outweighs the extra time investment.

Image courtesy of Marco Verch

Writing support from Sarah Sabir